Molecular Brain Research, 11 (1991) 79-87 ~) 1991 Elsevier Science Publishers B.V. 0169-328X/91/$03.50 ADONIS 0169328X91703208

BRESM 70320

79

The differential effects of cell density and NGF on the expression of tyrosine hydroxylase and dopamine fl-hydroxylase in PC12 cells

H.C. Badoyannis ~, S.C. Sharma ~ and E.L. Sabban 2 Department of Cell Biology and Anatomy and 2Department of Biochemistry and Molecular Biology, New York Medical College, Valhalla, NY

10595 (U.S.A.)

(Accepted 19 March 1991)

Key words: Tyro~me hydmxylase; Dopamine ~hydroxylase; Cell density; Nerve growth factor; PC12 cell; mRNA; Dopamine; Norepinephrine; Development; Cell aggregation

Expression of neuretransmitter phenotype during development of the nervous system is determined by several micro-environmental factors including cell aggregation. In order to delineate the role of cell aggregation and nerve growth factor (NGF) in regulating ~techolamine expression, tyrosi~ hydroxylase (TEl) and dopamine ~hydroxylase (DBED mRNA levels were examined in PC12 cells at different cell densities with and without L~GF treatment. Upon plating of PC12cells from low density (0.3--1.0 x 10 s ce~ls~Cl]ll 2) to high density (0.5-2.0 × 10 e cells/cm 2) TH mRNA levels increased &fold w i t ~ t day and remained at this level for several days. In cells replated from high to low density, TH mRNA returned to o~t~al ieve~ within 1 day. In addition to TH mRNA, TH protein and dopamine levels were also found to increase in high-density cultures. In contrast to the increase in TH mRNA, DBH mRNA decreased about 40% in cells plated from low to high density. Hence, cell density differentially regulated TH and DBH mRNA levels. Unlike cell density, NGF treatment led to a decrease in bo~ TH and DBH mRNA levels. However, when NGF treated cells were replated from low to high density, TH and dopamine levels increased. Thus NGF did not alter the density dependent regulation of TH. Similarly, TH mRNA levels increased in F4 cells, a mutant PC12 cell line unresponsive to NGF, when plated from low to high density. DBH mRNA decreased to undetectable levels when NGF treated PCI2 cells were plated to high density, demonstrating a synergefic effect of cell density and NGF treatment on DBH mRNA levels.

INTRODUCTION

Extracellular determinants which influence the expres- sion of neuronal characteristics include trophic factors and/or hormones as well as cell-cell contact. Determina- tion of neurotransmitter phenotype during development often coincides with the process of cellular aggregation 4" 16,40,s2 implicating a role for cell-cell contact in the

specification of transmitter phenotype. Farther evidence for the role of cell-cell contact in influencing neurotrans- mitter expression is evident from in vitro studies where alterations in cell density (i.e. extent of cell contact) resulted in changes in neurotransmitter traits 4"19~9~.

One of the better understood transmitter systems is that of the catecholamines. Tyrosine hydroxylase (TH), the first and rate firmting enzyme in the catecholamine biosynthe~c pathway, is induced in high density cultures of adrenal chromaffin cells or non-NGF-treated PC12 cells 1'19"2"/'29'44"50"56. The induction of TH specific activity

was shown not to result from diffusible factors 1"44, and

could be mimicked in bovine adrenal chromaffin cells by the addition of isolated plasma membranes, suggesting

that the molecule(s) that mediate TH induction are locali_zed within the membranes 56. Recently, a corre- sponding induction of TH m R N A has also been de- monstrated 2°'55'5~. The induction of TH activity can be prevented by transcriptional inhibitors ~,ss,57 indicating

that TH is probably regulated by cell-cell contact at the transcriptional level.

The studies on the induction of TH by cell density have concentrated on calf chromaffin cells or naive PC12 cells, a rat pheochromocytoma cell line. These chromaffin cells do not extend neurite-like processes when treated with NGI ~ . When PC12 cells are differentiated into a neuronal-like phenotype, it is unknown whether TH is still Susceptible to induction by cell density. The effects of cell density on the catecholaminergic phenotype of PC12 cells treated with NGF were evaluated in this study. NGF is an important component necessary for the survival and differenfiatiGn of catecholaminergic expres~,- ing sympathetic neurons both in ~Avo and in vitro ~2"ts'63.

Short-term treatment of PC12 cells with NGF for several minutes, led to phosphorylation of TH 3°'4" and to increased enzymatic activity 28. A transient 2-fold induc-

Correspondence: E.L. Sabban, Department of Biochemist~ and Molecular Biolog~;,F~:ew Yo:~ Medical College, Valhalla, NY 10595, U.S.A. Fax: (l) 914 9934058.

80

tion in TH m R N A was observed upon t reatment of PCI2

cells with N G F for 4 h 42 and recently this short- term

induction has been shown to proceed via transcriptional

activatiott e f the T H gene, maximal at 1 h of t reatment ,

and the specific transcriptional regulatory site has been

localized and shown to be media ted i,y a complex that

contains c-fos 23. Longer- term treatment of PC12 cells

with concentrations of N G F that lead to neuri te exten-

sion decreased the specific activity of both T H and

dopamine /~-hydroxylase (DBH) , the enzyme which

catalyzes the biosynthesis of norepinephrine from dopa-

mine. However, since the cell size increased by a

comparable amount the activity per cell was un-

changed 29. A decrease of about two fold in the relative

levels of TH m R N A in PC12 cells following several days

t rea tment with N G F has been reported 42.

Many of the t reatments which induce the activity of

T H have also been shown to induce ac*~ivity of DBH.

Therefore it was deemed important to see whether

regulation of D B H m R N A paral leled that of T H m R N A

by cell density and N G E

Results presented here illustrate that PC12 cell density

differentially regulates T H and D B H m R N A levels.

These density dependent changes were accompanied by

alterations in overall catecholaminergic phenotypic ex-

pression as assessed by protein and catecholamine levels.

PC12 cells differentiated along the neuronal-l ike pathway

were equally susceptible to induction of 7"1-I by increased

cell density. The densi ty-dependent regulation of D B H

m R N A was even more pronounced in NGF-treated cells,

indicating that cell density and N G F have a synergetic

effect on the regulation of DBH.

MATERIALS AND METHODS

Cell growth and treatme~; PC12 cells 29 and F~ cells, a spontaneously arising mutant of the

PC12 cell line 46, ,~re obtained from Dr. Lloyd Greene, Columbia Univel~ity. PCI2 cells were grown as previously described 5s on c~llagen coated Falcon tissue culture dishes in RPMI 1640 medium (Gibco) sup~Jlemented with 10% heat-inactivated horse serum, 5% fetal boviw~ serum (Hazelton) and 50 IU/ml penicillin and 50 #g/ml streptomycin. Cells were maintained at 37 °C in a humidfied, 8% CO 2 atmosphere. Cell cout~:~ were performed using a hemocyto- meter and c¢H viability (~90%) determined by Trypan blue exclusion. F4 cells were grown as described for PC12 cells except that plastic tissue culture dishes without collagen were used.

For cell density experiments, PC12 cells were triturated and plated to !o,~ density (0.3-1.0 × l0 s cells/cm2), and were allowed to grow for 2-3 days. In order to reduce cellular aggregation in these low density cultures, cells were treated with 1 mM EGTA (6-12 h) prior to trituration as described by Green et al. 2s. EGTA treatment did not affect cell viability or mRNA levels for TH and DBH. The cells were then replated to the same low density or to a high density (0.5-2.0 × 106 cells/cm2). For cell density experiments with F4 cells, cells were harvested by incubating cells in 0.1% trypsin in sterile PBS for 2-3 min ~nd plated tc either low or a 36-fold higher density. Since F4 cells have a f~tened, fibroblast-like morphology, this 36-fold difference in plating density w~.'s used to achieve low-density

cultures with little cell-cell contact similar to that of low-density PC12 cultures.

For NGF treatment, PC12 cells were incubated for various times with either 7S NGF (Sigma) at a final concentration of 300 ng/ml or /~-NGF (University of Michigan-Flint) at a final concentration of 10O ng/ml.

RNA isolation and Northern atm.lyjis Total RNA was isolate d _ from PC12 cell cultures according to the

method of llaria et al. 32. RNA concentrations were determined by absorbance readings at 260 nm. Total RNA was fractionated on i.3% agarose gels in the presence of 2.2 M formaldehyde 1~ and transferred to nitrocellulose filters 62. Filters were baked in a vacuum ov~E at 80 °C for 2 h. Prehybridization was at 42 °C in 50% iormamide, 5 x SSC (20 × concentrated: 3 M NaCi and 0.3 M sodium citrate, pH 7), 5 × Denhardt's solution (50 × Denhardt's : 1% polyvinylpyrrolidone) (Sigma), Ficoll 400 (Pharmacia), and bovine serum albumin (Pharmacia), 50 mM sodium phosphate pH 7.4, 0.1% SDS, and 100/~g/ml salmon sperm DNA (Sigma) 4s. Hybridization was carried out in the same solution overnight with one of 3 probes labelled by the random primer method 21 using the Multiprime Kit (Amersham). [a-SZP]dCTP (>3000 Ci/mmol) from New England Nuclear was used to label probes for TH, DBH or 18S rRNA to a specific activity of approximately I x 109 dpm//~g.

The TH probe was a 1.1 kb 5" EcoRl eDNA fragment to rat T n 42 kindly provided by Dr. Edward Ziff, New York University Medical Center. The DBH probe was a cDNA corresponding to nucleotide 1297 to the 3" end of the 2.7 kb mRNA species for rat DBH cloned by McMahon et al. 47 and the 18S rRNA DNA probe ~ was a gift from Dr. I. Wool, University of Chicago.

Filters were washed 3 times for 10 min at room temperature in 2 × SSC, 0.1% SDS and 3 times for 45 rain at 62 °C in 0.1% SSC, 0.1% SDS. Filters were dried and then exposed to Kodak XAR film at -70 °C, with intensifying screens. Filters were stripped of probe between hybridizations by placing them in boiling water and then allowing to cool to room temperature.

The hybridization signals obtained for 18S rRNA were used to normalize the relative amounts of TH and DBH mRNA since the hybridization signal obtained with 18S rRNA was found to be linear with RNA concentration (McMahon and Sabban, in preparation). Hybridization signals were quantitated by performing densitometric scans of autoradiograms in the range linear with concentration using a Hoefer GS 300 TransmittancefReflectance scanning densitometer.

DNA analysis To be certain that rRNA levels were not affected by cell density,

DNA levels of PC12 cell samples were analyzed according to the method of Labarca and Paigen sT. Fluorometry employing Hoescht DNA binding dye and using a TKO 100 mini fluorometer (Hoefer Scientific Instruments) revealed that the rRNA/DNA ratio did not vary between low and high density cultures.

Analysis of cellular catecholamine levels Cell samples were suspended in 0.4 M perciflodc acid, 0.1 mM

EDTA, 7.6 mM sodium bisulfite and subjected to 3 freeze-thaw cycles. Following centrifugation, the amount of dopamine and norepinephrine in cell supernatants was determined by I-IPLC using electrochemical detection (BAS, LC-4B~. as previously described 3s. The catecholamines were normalized to cellular protein from the perchloric acid-insoluble pellets with the BioRad Protein Assay Kit s .

Western blot analysis Cell samples were suspended in PBS, and a!iquots taken for

protein determination s and the remainder centrifuged and suso pended in gel buffer (10% gly~rol, 5% SDS, 100 ~g/ml pyronin Y dye in 0.16 M Tris-phosphate buffer pH 6.7) and sonicated (Branson Sonifier 250). Protein was fractionated on a 10% SDS polyacryl- amide gel according to ~he method of Laemmfi 3s and transferred to nitrocellulose (SBD-1000 ABN Polyblot transfer apparatus, Amer-

81

ican Bionetics) according to the method of Kyhse-Andersen 36. Filters were blocked in PBS, 1% Tween 20, and 0.5% dry milk, followed by overnight incubation at 4 °C with anfisera to TH (1:250, Eugene Tech). An anti-lgG conjugated to alkaline phosphatase (1:7500, Promega) was employed for the visualization of immuno- reactive protein. Densitometry was used for quantitation.

RESULTS

Effects of cell density on TH and DBH mRNA levels in PC12 cell cultures

To determine whether the expression of D B H is subject to regulation by cell density, both TH and D B H mRNAs were examined in PC12 cell cultures grown at low (0.3-1.0 x 10 ~ cells/cm 2) and high density (0.5-2.0 x 106 cells/cm2). PC12 cells grown at low density for 3 days

Low High density density

A

T H ;! ~ --- 1.9 Kb

B

DBH - - 2.7Kb "" 2.5Kb

C

18S rRNA

~

Fig. 1. Comparison of mRNA levels for TH and DBH in low- and high-density PC12 cell cultures. Total cellular RNA (15/tg per lane) from PC12 cell cuitures grown initially for 3 days at low den~ty (approximately 5 x 104 cells/era ~) and then replated and grown for 3 days at either the same low density (duplicate samples) or 12-fold higher density (tripficate samples) was subjected to Northern blot analysis and hybridized with probes for: (A) TH mRNA (13) DBH mRNA and (C) 18S rRNA. The third low density sample is not shown in C since it was degraded.

were replated and grown for 3 days at either the same low density or a I2-foid higher density. Total R N A from these cultures was analyzed by Northern blots for levels of TH and DBH mR2.4A. Hybridization with a cDNA probe for TH identified a single m R N A species of 1.9 kb (Fig. 1A) as previously reported 39"43. A 4-fold increase in TH mRNA levels was observed when cells were grown at high density. Similar results were obtained from 8 separate experir~ents each with triplicate samples (1.0 + 0.06 at low density, 3.88 + 0.45 at high density, mean + S.E.M.; P < 0.001 by t-test). In this study, ,~,alues for TH and DBH m R N A s were normalized to 18S rRNA (e.g. Fig. 1C). The signal obtained for 18S rRNA is an appropriate control since the hybridization was finear. with R N A concentration and the ratio of 18S rRNA to D N A did not change with cell density (data not shown).

As previously reported, hybridization for DBH in PC12 cells detected two m R N A species of 2.5 and 2.7 kb (Fig. 1B) probably arising from differential poly- adenylation 47. In contrast to the increase in TH m R N A levels observed in high-density cultures, D B H m R N A levels decreased an average of approximately 40% (1.0 + 0.05 at low density, 0.58 +_ 0.05 at high density, mean + S.E.M.; P < 0.001). This result was based on the sum of both D B H m R N A species. Examining the ratio of the 2.7 kb to the 2.5 kb m R N A species revealed that the ratio changed from 0.56 +_ 0.04 in low density cultures to 0.34 + 0.04 (mean +_ S.E.M., n = 16) in high-density cultures implying that there was differential regulation of the two

DBH mRNAs. The time course of the density dependent changes in

20 . . . _ - - - - e

g

L~

~..~"~" 5O!o ~

01| i i i i i ,

0 12 24 3fi 48 60 72 Hours of culLure

Fig. 2. The tim© course of density-dependent changes in TH mRNA levels in PC12 cell cultures. PC12 ceils growing at high density (~1.2 × 106 cells/cm 2) were harvested (0 h time point) and replated to a 12-fold lower density (-1 × 105 cells/cm2). After 24 h, duplicate cultures were replated (indicated by arroyo) at either the same low density (open circles) or to a 20-fold higher density (-2 x 106 ceils/cm2; closed circles). The cells were harvested at the indicated C-,me. poin~ and TH mRNA levels were determined from densito- metric scans of Northern blots and nonn~zed to 18S rRNA. Each symbol represents a single culture.

82

2i0

c~ 180 ~ e '~50

Z' replate ~' 120 ~

g ¢. 90 oO°P °mine

30 ~ • . C, m'~ :~ - - . - -

Norepmephr, ne . . ~ .

0 12 24 36 48 60 72 Houvs of c'JItuee

Fig. 3. The time course of-density-dependent changes in cellular dopamine and norepinephrine content in PC12 cell cultures. Aliquots taken from PC12 cells grown and harvested as previously described (in legend of Fig. 2) were analyzed for cellular dopamine (circles) or norepinephrine (diamonds) content. Arrow indicates time of replating to either low density (ope, symbols) or high density (closed symbols). Each symbol represents a single culture.

1.4

~= 1.2 >- .~ t.o

W o.e

~ 0.4

~ o2

o.o I I I | I I t I

0 ! 2 3 4 5 6 7 Daye of cu]tuve with NGF

o = TH o = DBH

O 1 - - - - O

" 0 . . . . . . . . . . . }

I I I

8 9 10

Fig. 5. Time course of NGF-mediated changes in mRNA levels for TH and DBH. PCI2 cells were plated to low density and allowed to adjust to that density for 1 day and then treated for various times up to 10 days with 7S NGF (300 ng/ml) or/~-NGF (100 rig/nil). TH mRNA (circles) and DBH mRNA (diamonds) levels were normal- ized to 18S rRNA. Since the results were similar with 7S or/~-NGF, the results were combined and the mean and S.D~ plotted.

TH m R N A and catecholamine levels were analyzed in

PC12 cultures at various densities. As shown in Fig. 2,

TH m R N A levels in cells plated from high to a 12-fold

lower density had decreased by 72% after 24 h and

remained essentially unchanged when replated to the

same low density. Cells replated to a 20-fold higher

density demonstrated an 18-fold increase in T H m R N A 1 and 2 clays later.

Analysis of catecholamine content (Fig. 3) in these

cells revealed that dopamine levels decreased by 48%

within 24 h of replating from high to low density and

remained at comparable levels when replated to the same

low density. However, dopamine levels increased 4.9-

fold in cultures replated to a 20-fold higher density. This

Low d e n s i t y High d e n s i t y J L A

HD ~ ~, r CtZ'I C t r l l d 2d l d 2d 2d(15) 2d( lO)

- 6 9 gd

- 4 6

- 3 0

Fig. 4. Western blot analysis of TH protein levels in low- and high-density PC12 cell cultures. High-density PC12 cells (HD Ctrl; -~1 × 10 e celis/cm 2) were plated to low density ~-8 × 104 cells/cm2), grown for 2~days and harvested (Low density Ctrl), then repl~ted to the same low density or to a 12-fold higher density and examined one day (ld) or two days (2d) later. The same amount of total protei:l (30 #g) was used for all lanes except the last two which had 15/~g (2d(15)) and 10/~g (2d(10)), respectively.

TH

LD HD

i:iii!i!ii!!i! !i i!i i DBH

tBS PRNA

Fig. 6. Density-de~ndeat regulation of TH and DBH mRNA levels in control and NGF-treated PCI2 cells. PC12 cells were grown at low density (--1 × l0 s cells/cm 2) for 4 days in the presence of/~-NGF (100 ng/ml) and then replated to the same low density (LD) or a 12-_fold higher density (HD) and further grown for 2 days. Total RNA (10 #g) from duplicate cultures was subjected to Northern analysis and hybridized with probes for: TH, DBH, and 18S rRNA.

100 e--

8o

~ 40' s

~Dopamine

[~Noreptnephrtne

--'1"-

Low den~-~Lv High den~=ity

Fig. 7. Density-dependent changes in dopamine and norepinephrine content in NGF-treated PCI2 cells. Aliquots taken from triplicate cultures of PCI2 cells initially treated with NGF for 4 days then replated to low density or high density and grown for 2 days in the absence of NGF were analyzed for cellular dopamine and norepi- nephrine content. Results are based on o~ experiment with triplicate cultures (mean + S.E.M.).

result is consistent with the observed indu~:tion of TH mRNA (Fig. 2). In the same cell culture~, norepineph- fine levels decreased 31% when examined 24 h after plating from high to low density and remained at similar levels in cultures replated either to the same low or ~0-fold higher density.

The relative amount of TH protein expressed in low and high-density cultures was examined by immunoblots In order to determine whether there were corresponding density dependent changes in TH protein levels. As expected, a single immunoreactive polypeptide of ap- proximately 60 Kd was observed in Western blots for TH

F4 cells Low Hig h

density dens,zy

T H

Fig. 8. Comparison of mRNA levels for "It/between low- and high-density F4 cell cultures. Northern blot analysis of TH mRNA from F4 cells at low and high density. Total cellular RNA (20/~g per lane) from F4 cell cultures Flared and grown for 2 days at either low density or at a 36-fold higher density (as described in the experimental procedures) were su~,jec~ed to Northern analysis and probed for TH mRNA (triplicate samples).

83

(Fig. 4). High density cultures replated to a 12-fold lower density showed a decrease in TH protein levels when examined 2 days later (Fig. 4). When these low density cells were replated to high density, TH protein levels were elevated more than 3-fold. This increase in TH protein was apparent after 1 day at high density. In two other experiments, as much as a 10-fold increase in TH protein was observed at high density. Thus, increased cell density led to increased levels of TH mRNA, TH prote~n, and a corresponding increase in dopamine levels in PC!2 cultures.

Effects of cell density on NGF-treated PC12 cells We examined the effects of cell density on TH and

DBH expression in NGF treated PC12 cell cultures (Fig. 5). Cells growing at either low or medium density were treated for various times , p to 10 days with either 7S or /~NGF. The PC12 cells elaborated neuronal-like processes t~at were visible after 3 days of NGF exposure, and became more extensive after 4 to 10 days of continuous NGF exposure. NGF treatment for 1 day led to a slight induction of TH mRNA levels (14%), however, with longer exposures to NGF, TH mRNA decreased. In 4 separate experiments TH mRNA levels decreased an average of 37% when observed after 4 days of NGF treatment (1.0 +_ 0,05 in control cultures, 0.63 _+ 0.05 in NGF-treated cultures, mean _ S.E.M.; P < 0.001).

DBH mRNA levels decreased rapidly with NGF treatment (Fig. 5). Averaging from 4 separate experi- ments, DBH mRNA levels were reduced to 33% of the control values in cultures treated with NGF for 4 days (1.0 _+ 0.09 in control cultures, 0.33 _+ 0.06 in NGF- treated cultures, mean -+ S.E.M.; P < 0.001). In both cases TH and DBH mRNA remained at reduced levels for up to 10 days of NGF treatment (Fig. 5). Similar decreases in both TH and DBH mRNA levels with NGF treatment were observed regardless of cell density.

In order to ~certain whether TH and DBH mRNA levels would return to control values upon withdrawal of NGF, PC12 cells were treated for 4 days with NGF, the time at which TH and DBH mRNA levels were maxi- mally reduced (Fig. 5). NGF treatment was discontinued and cells further incubated for 4 days in media lacking NGE The neuronal processes withdrew, yet TH and DBH mRNA levels still remained reduced at these times

(data not shown). To examine the effect of ceil density on TH and DBH

mRNA levels in NGF treated PC12 cells, ceils were grown at low density without NGF (control cultures) or in the presence of NGF for 4 days and subsequently replated to low and hi?,h densities. Northern blot analysis (Fig. 6) revealed an induction of TH mRNA levels when replated to high density in NGF treated high density

84

cells. DBH mRNA levels were ma ,<edly reduced to undetectable levels when NGF-tre~ e d cells were grown at high density. At high density r jBH mRNA was also undetectable in overexposed :~utoradiograms. Similar results were obtained whet~er NGF treatme ,~ was continued or discontinued at ~he time of replating. The magnitude of the induction in TH mRNA was similar in both control and NGF treated PC12 cultures.

Analysis of cellular catecholamine levels (Fig. 7) revealed a 3.7-fold increase in dopamine levels with norepinephrine remaining at low ¢,~,centrations in NGF- treated PC12 cells plated to high density.

Effects of cell density on F4 cell cultures A cell line unresponsive to NGF was used to compare

~;:e effects of cell density to those observed in PC12 cells. The [:4 cell line is a clonally derived, spontaneously arisin3 mutant of PC12 cells that does not respond to NGI ~ . F4 cells have a flattened, fibroblast-like mor- pholog ~, and appear to have lost most of the chromaffin ceii ch,~racteristics. Although F4 cells have been reported to lack ~2etectable levels of "[H mRNA 42, we were able to detect an mRNA, albeit at lower levels relative to PC12 cells, thai hybridized with the TH cDNA which was indistinguh:hable froni TH mRNA in PC12 cells (Fig. 8). No detectable DBILi mRNA was observed. When plated to high densi: y, F4 cells showed a 2.~i-fold increase in TH mRNA compared to low density cultures. Thus, a non-NGF re~;ponsive PC12 variant still demonstrated density dependent regulation of TH.

DISCUSSION

The present study demonstrated that cell density differentially regulated TH and DBH mRNA levels in PC12 cells, whereas NGF alone had a comparable effect on both TH and DBH mRNA. Cells differentiated along the neural pathway also responded to regulation by cell density. There was a synergetJc effect of NGF and cell density on DBH mRNA levels.

Differential regulation of TH and DBH mRNA by cell density

At h~gh cell densities, TH mRNA and protein levels were markedly increased, and this increase was accom- panied by elevated cellular d o r o ~ a e levels. In contrast, DBH mRNA levels decline J, with a more pronounced reduc ion in the 2.7 that, m the 2.5 kb mRNA species. These results indicate :aat TH and DBH are differenti- ally regulated by co', density in PC12 cells.

Faere are a r Jmber of physiological conditions in which DBH acti lty is regulated in concert with TH. The enzymatic act: ~ty of TH and DBH are induced in vivo

t, nder situations such as exposure to cold and immobili- zation stress 64, glucocorticoids 13'14, preganglionic sympa- thetic nerve stimulation ~° and drugs such as reserpine ~ ' ~. Examination of some of these factors at the mRNA level has shown that DBH mRNA parallels the increase in TH mRNA, e.g. the induction of adrenal TH and DBH mRNAs following reserpine treatment 47.

There are however several studies that indicate that TH and DBH are not always regulated together during long-term treatments. Nestler et al. 51 showed that chronic treatment of rats with a variety of antidepressants led to a decrease in TH protein and mRNA levels in the locus coeruleus, the major adrenergic nucleus in the CNS, without affecting DBH mRNA levels. Similarly, following electrical preganglionic stimulation of the tho- racic ganglia in humans, a large induction of DBH mRNA (6-fold) occurred with a much smaller change in TH mRNA 59.

The present results with DBH mRNA are in contrast to an observed induction of DBH activity with increased cell density of bovine chromaffin cells 2. It is possible that changes in DBH mRNA are not correlate0 to changes in DBH activity, especially since DBH activity is well known to be affected by endogenous inhibitors. How- ever, since the decrease in DBH mRNA with NGF reported here correlates with the decrease in specific activity of DBH with NGF 29, this possibility is highly unlikely.

A differential regulation of TH and DBH mRNA in PC12 cells was observed following membrane depolari- zation with either elevated KC! or veratridine 34. In depolarized cells, TH mRNA was found to rise 2- to 5-fold, without a significant change in DBH mRNA levels. Thus, both depolarization and cell density have a differential effect on TH and DBH mRNA levels in PC12 cells. It is unlikely that these results simply reflect the inability of DBH mRNA to be induced in PC12 cells. DBH mRNA in PC12 cells was found to be induced in parallel with TH mRNA with other treatments such as glucocorticoids or cAMP analogs (McMahon and Sab- ban, in preparation). In addition: the present study illustrates that TH and DBH mRNAs are ~egulated coordinately by prolonged NGF treatment. These results suggest that the TH and DBH genes may contain some common and some unique regulatory elements.

Effect of cell density and NGF on TH and DBH The present results are consistent with the results of

Leonard et al. 42 who showed that prolonged NGF treatment (which leads to neurite outgrowth) led to about a 2-fold decrease in the specific amount of TH mRNA. Furthermore, we show that the amount of DBH mRNA decreased under the same conditions. Thus, clearly

neurite outgrowth of these cells can proceed without induction of TH or DBH. NGF treatment was also previously shown to enhance the post-translational pro- cessing of DBH in PCI2 cells 5s. High concentrations of NGF (20 times higher than needed for neurite outgrowth in PC12 cells) induced both TH and DBH activity in bovine chromaffin cells 3 and in dissociated sympathetic ganglia cultures 3~. In vivo, aciministration Of NGF led to increases in activities of both TH and DBH in the superior cervical ganglia of newborn rats 61.

The present study also shows that the induction of TH mRNA by cell density can proceed in PC12 cells differentiated into a neuronal-like phenotype. To our knowledge, this is the first reported induction of TH by cell density in neuronal-like cells. The effect of cell density has been studied in sympathetic neurons 4. How- ever, sympathetic neuronal cultures did not show altered TH levels per cell when grown at high density whereas at the same time there was a large increase in the levels of substance P and choline O-acetyltransferase per neuron 4. Membrane preparations from a variety of sources and particularly the dorsal root ganglia 5 have been effective in inducing the choline O-acetyltransferase activity 6'55'66.

The molecular component(s) involved in the density dependent regulation of TH and DBH expression in PC12 cells remains to be determinea. Saadat and Thoenen s6 have identified an integral membrane pro- tein(s) as being involved in the density dependent induction of TH in chromaffin ceils. In addition, protein factors that specify cell density directed changes in other neurotransmitter systems, such as substance P, have been characterized 6,66.

The second messengers responsible for the cell-ceU contact effect on the catecholamine biosynthetic enzymes also have yet to be elucidated. Although transcriptional activation is likely to be responsible for the increase in TH with increased cell density 9, Fader and Lewis 2° showed that the density dependent regulation of TH in PC12 cells is not mediated by the first 773 nucleotides of the 5" flanking region of the TH gene which include cis-elements regulated by cAMP. Interestingly, they found an interac, tion between the effect of cell contact and cAMP o~ elevation of TH mRNA with a smaller induction by forskolin at higher cell density z°.

Particularly intriguing in the present study are the findings of the interaction between NGF and cell density on the regulation of DBH mRNA. Treatment with NGF did no~. ~lter the responsiveness of TH mRNA in PC12

REFERENCES

1 Acheson, A.L. and Thoenen, H., Cell contact-mediated regu- lation of tyrosine hydroxylase synthesis in cultured bovine adrenal chromaffin cells, J. Cell Biol., 97 (1983) 925-928.

85

cells to cell density. Whereas, the combination of increased cell density and NGF had a synergetic effect on DBH mRNA. In this regard, it is perhaps pertinent to point out that NGF treatment increased the rates of cell-substratum and cell-ceU adhesion of PC12 cells c'° and NGF can alter concentrations of cell adhesion molecules such as N-CAM 54 and Ng-CAM zz.

Cell-cell contact is known to be an important mech- anism in tissue formation, cytodifferentiation and cell growth. The importance of cell-cell contact in develop- ment has been suggested by elegant experiments with hepatocytes 24'49. In these instances plasma membranes from various tissues were able to mimic the effect of cell-ceU contact on the maintenance of liver-specific proteins 49. This indicated that the cell surface modulators did not induce new expression of dormant genes but rather regulated phenotypic expression of genes already switched on during terminal differentiation. It has been shown in hepatocytes that this cell contact regulation of phenotypic expressioa of liver proteins involves transcrip- tional regulation TM. "In the nervous system, the develop- ment and maintenaace of function relies on the interplay of many intrinsic and extrinsic factors. It has been suggested that cellular contact or aggregation may pro- vide the stimulus for expression or induction of catechol- aminergic traits since they appear simultaneously with the aggregation of neuroblasts to form primitive sympathetic ganglia 16. Cell-cell contact has been shown to be an important factor in the maintenance of TH in chromaffin cells 1'5s'56. Early neural crest cells are multipotent and

can give rise to various cell types, including sympathetic and chromaffin cells. The sympathoadrenal precursor is believed to differentiate along the chromaffin pathway or the neuronal pathway depending on the presence of glucocorticoids or NGF respectively 7'1a'~'53. Therefore

the interplay of these factors with cell-cell contact should allow a more detailed understanding of their role in development and in neurotransmitter expression.

Acknowledgements. We gratefully thank Dr. Anne McMahon for her generous assistance and excellent suggestions. We thank Dr. Lloyd Greene (Columbia University) for the PC12 and F4 cells, Dr. Edward Ziff (NYU Medical Center) for the cDNA clone to TH, and Dr. Ira Wool (University of Chicago) for the clone of the 18S rRNA gene. We also wish to thank Drs. Stuart Newman and Andrew Powers (New York Medical College) for careful reading of the manuscript and Edward Kilbourne for usef~al suggestions. This work was supported by NIH Grant N28869, NIH Research Career Development Award NS01121 (to E.L.S.), and partially supported by research grants sponsored by the Hellenic University Club and Medical Society of New York (to H.B.).

2 Acheson, A., Edgar, D., Timpl, R. and Thoenen, H., Laminin increases both levels and activity of tyrosiee hydroxyiase in calf adrenal chromaffin cells, J. Cell Biol., 102 (1986) 151-159.

3 Acheson, A.L., Naujoks, K. and Thoenen, H.~ Nerve growth factor-mediated enzyme induction in primary cultures of bovine

86

adrenal chromaffin cells: specificity and level of regulation, ./. Neurosci., 4 (1984) 1771-1780.

4 Adler, J.E. and Black, I.B., Sympathetic neuron density differentially regulates transmitter pl~enotypic expression in culture, Proc. Natl. Acad. Sci. U.S.A., 82 (1985) 4296-4300.

5 Adler, J.E. and Black, I.B., Membrane contact regulates transmitter phenotypic expression, Dev. Brain Res., 30 (1986) 37-241.

6 Adler, .I.E., Schleifer, L.S. and Black, I.B., Partial ~rification and characterization of a membrane-derived factor ~egulating neurotransmitter phenotypic expression, Proc. Nad. Acad. Sci. U.S.A., 86 (1989) 1080-1083.

7 Black, I.B., Adler, J.E., Bohn, M.C., Jonakait, G.M., Kessler, J.A. and Markey, K.A., Transmitter phenotypic plasticity in developing and mature neurons in vivo. In I.B. Black (Ed.), Cellular and Molecular Biology of Neuronal Development, Plenum, New York, 1984, pp. 117-130.

8 Bradford, M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the plincip|es ~t" protein-dye binding, An,i. Biochem., 72 (197¢~) ;~45.

9 Carlson, C.D. and Tank, A.W., Regulatio~ of tyrosine hydroxy- lase gene transcription in PCI8 cells by cell density, Soc. Neurosci. Abstr., 16 (1990)~294.

10 Chalazonitis~ A., Rice, P.J. and Zigmond, R.E., Increased gan~iionic tyrosine hydroxylase : ld dopamine beta hydroxylase activities following preganglionic nerve stimulation: role of nicotinic receptors, J. P,~armacoi. Exp. Ther., 213 (1980) 139-143.

11 Chan, Y.-L., Gutell, R., Noller, H.F. and Wool, I.G., The nucleotide sequence of a rat 18 S ribosomal ribonucleic acid gene and a proposal for the secondary structure of 18 S ribosomal ribonucleic acid, J. Biol. Chem., 259 (1984) 224-230.

i2 Chun, L.L. and Patterson, EH., Role of nerve growth factor in the development of rat sympathetic neurons in vitro. I. Survival, growth and differentiation of catecholamine production, J. Cell Biol., 75 (1977) 694-704.

13 Ciaranello, R.D., Wooten, G.F. and Axelrod, J., Regulation of dopamine beta hydroxylase in rat adrenal glands, J. Biol. Chem., 50 (1975) 3204-3211.

14 Ciaranello, R.D., Wooten, G.E and Axelrod, J., Regulation of :at adrenal dopamine beta hydroxylase. II. Receptor interaction in the regulation of enzyme synthesis and degradation, Brain Res., 113 (1976) 349-362.

15 Clayton, D.E, Harrelson, A.L. and Darnell, Jr., J.E., Depen- dence of liver.specific transcription on tissue organization, Mol. Cell Biol., 5 (1985) 2623-2632.

16 Cochard, E, Goldstein, M. and Black, I.B., Initial development of the noradrenergic phenotype in autonomic neuroblasts of the rat embryo in vivo, Dev. Biol., 71 (1979) 100-114.

17 Davis, L.G., Dibner, M.D. and Battey, J.E, Basic Methods in Molecular Biology, Elsevier, New York, 1986.

18 Doupe, A.J., Landis, S.C. and Patterson, EH. Environmental influences in the development of neural crest derivatives: glucocorticoids, growth factors and chromaffin plasticity, ./. Neurosci., 5 (1985) 2119-2142.

19 Edgar, D.H. and Thoenen, H., Selective enzyme induction in a nerve growth factor-responsive pheochromocytoma cell line (PC12), Brain Res., 154 (1978) 186-190.

20 Fader, D. and Lewis, E.J., Interaction of cyclic AMP and cell-cell contact in the control of tyrosine hydroxylase RNA, Moi. Brain Res., 8 (1990) 25-29.

21 Feinberg, A.E and Vogelstein, B., A technique for radiolabel- ling DNA restriction endonuclease fragments to high specific activity, Addendum Anal. Biochem., 137 (1984) 266-267.

22 Friedlander, D.R., Grumet, M. and Edelman, G.M., Nerve growth factor enhances expression of neuron-~ia cell adhesion molecule in PC12 cells, J. Cell Biol., 102 (1986) 413-419.

23 Gizang-Ginsberg, E. and Ziff, E.B., Nerve growth factor regulates tyrosine hydroxylase gene transcription through a nucleoprotein cor~plex that contains c-Fos, Genes Dev., 4 (1990)

477-491. 24 Grayson, D.R., Costa, R.H. and D~rnell, J.E., Regulation of

hepatocyte-specific gene expression, Ann. N.Y. Acad. Sci., 557 (1989) 243-255.

25 Green, S.H., Rydel, R.E., Connolly, J.L. and Greene, L.A., PC12 cell mutants that possess low- but not high-affmity nerve growth factor receptors neither respond to nor internalize nerve growth factor, Z Cell Biol., 102 (1986) 830-843.

26 Greene, L.A., Aletta, J.M., Rukenstein, A. and Green, S.H., PC12 pheochromocytoma cells: culture, nerve growth factor treatment, and experimental exploitation, Methods Enzymol., 147 (1987) 207-216.

27 Greene, L.A. and Rein, G., Synthesis, storage and release of acetylcholine by a noradrenergic pheochromocytoma cell fine, Nature, 268 (1977) 249-351.

28 Greene, L.A., Seeley, P.J., Rukenstein, A., DiPiazza, M. and Howard, A., Rapid activation of tyrosine hydroxylase in response to nerve growth factor, .l. Neurochem., 42 (1984) 1728-1734.

29 Greene, L.A. and Tiscbter, A.S. Establishment of a noradren- ergic clonal l~ne of rat adrenal pheochromocytoma cells which respond to nerve growth factor, Proc. Natl. Acad. Sci. U.S.A., 73 (1976) 2424-2428.

30 Halegoua, S. and Patrick, J., Nerve growth factor mediates phesphorylation of specific proteins, Cell, 22 (1980) 571-581.

31 Hefti, E, Gnahn, H., Schwab, M.E. and Thoenen, H., Induction of tyrosine hydroxylase by nerve growth factor and elevated K + concentrations in cultures of dissociated sympa- thetic neurons, J. Neurosci., 11 (1982) 1554-1566.

32 llaria, R., Wines, D., Pardue, S., Jamison, S., Ojeda, S.R., Snider, J. and Morrison, M.R., A rapid microprocedure for isolating RNA from multiple samples of human and rat brain, ./. Neurosci. Methods, 15 (1985) 165-174.

33 Kessler, J.A., Corm, G. and Hatcher, V.B., Isolated plasma membranes regulate neurotransmitter expression and facilitate effects of a soluble brain cholinergic factor, Proc. Natl. Acad. $ci. U.S.A., 83 (1986) 3528-3532.

34 Kiibourne, E.J. and Sabban, E.L., Differential effect of mem- brane depolarization on levels of tyrosine hydroxylase and dopamine/~-hydroxylase mRNAs in PC12 pheochromocytoma cells, Mol. Brain Res.~ 8 (1990) 121-127.

35 Kuhn, L.J., Hadman, M. and Sabban, E.L., Effect of monensin on synthesis, post-translational processing, and secretion of dopamine beta hydroxylase from PC12 pheochromocytoma cells, Z Biol. Chem., 261 (1986) 3816--3825.

36 Kyhse-Andersen, J., Electroblotting of multiple gels: a simple apparatus without buffer tank for rapid transfer of proteins from polyacrylamide to nitrocellulose, J. Biochem. Biophys. Methods, 10 (1984) 203-209.

37 LaBarca, C. and Paigen, K., A simple, rapid and sensitive DIqA assay procedure, Anal. Biochem., 102 (1980) 344--352.

38 Laemmli, U.. Cleavage of structural proteins during assembly of the head of bacteriophage 1"4, Nature, 227 (1970) 680-685.

39 Lamouroux, A., Faucon Biguet, N., Samolyk, D., Privat, A., Salomon, J.C., Pujol, J.F. and Mallet, J., Identification of cDNA clones coding for rat tyrosine hydroxylase antigen, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 3881-3885.

40 LeDouarin, N.M., The Neural Crest, Cambridge University Press, Cambridge, 1982.

41 Lee, K.Y., Seeley, J., Muller, T.H., Helmer-Matyjek, E., Sabban, E., Goldstein, M. and Greene, L.A., Regulation of tyrosine hydroxylase phosphorylation in PC12 pheochromocy. toma cells by elevated K + and nerve growth factor, evidence for different mechanisms of action, Mol. Pharm., 28 (1985) 220- 228.

42 Leonard, D.G.B., Ziff, E.B. and Greene, L.A., Identification a,d characterization of mRNAs regulated by nerve growth factor in PC12 cells, Mol. Cell Biol., 7 (1987) 3156--3267.

43 Lewis, E.J., Tank, A.W., Werner, N. and Chikaraishi, D.M., Regulation of tyrosine hydroxylase mRNA by glucocorticoid and

cyclic AMP in rat pheochromocytoma cell line. Isolation of a eDNA clone for tyrosine hydroxylasc mRNA, J. Biol. Chem., 258 (1983) 14632-14637.

44 Lucas, C.A., Edgar, D. and Thoenen, H., Regulation of tyrosine hydroxylase and choline acetyitransferase activities by cell density in the PC12 rat pheochromocytoma donal cell line, Exp. Cell Res., 121 (1979)79-86.

45 Man~atis, T., Fritsch, E.E and Sambrook, J., Molecular Cloning, a Laboratory Manual, Cold Spring Harbor Press, New York, 1982.

46 McGuire, J.C. and Greene, L.A., Rapid stimulation by nerve growth factor of amino acid uptake by clonal PC12 pheochro- mocytoma cells, I. Biol. Chem., 254 (1979) 3362-3367.

47 McMahon, A., Geertman, R. and Sabban, E.L., Rat dopamine fl-hydroxylase: molecular cloning and characterization of the eDNA and regulation of the mRNA by reserpine, I. Neurosci. Res., 25 (1990) 395-404.

48 Molinoff, P.B., Brimijoin, S. and Axelrod, J., Induction of dopamine beta hydroxylase and tyrosine hydroxylase in rat hearts and sympathetic ganglia, J. Pharnmcol. Exp. Ther., 182 (1972) 1!6-129.

49 Nakamura, T., Yoshimoto, K., Nakauama, Y., Tomita, Y. and Ichihara, A., R°-ciprocal modulation of growth and differen- tiated functioas of mature hepatocytes in primary culture by cell-cell contact and cell membranes, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 7229-7233.

50 Naujoks, K.W., Korsching, S., Rohrer, H. and Thoenen, H., Nerve growth factor-mediated induction of tyrosine hydroxylase and of neurite outgrowth in cultures of bovine adrenal chromaf- fin cells: dependence on developmental stage, DeJ.,. Biol., 92 (1982) 365-379.

51 Nestler, E.J., McMahon, A., Sabban, E.L., Tallman, J.F. and Duman, R.S., Chronic antidepressant administration decreases the expression of tyrosine hydroxylase in the rat locus coeruleus, Proc. Natl. Acad. Sci. U.S.A., 87 (1990) 7522-~]526.

52 Patterson, P.H., Environmental determination of autonomic neurotransttdtter functions, Annu. Rev. Neurosci., 1 (1978) 1-17.

53 Patterson, P.H., Control of cell fate in a vertebrate neurogenic lineage, Cell, 62 (1990) 1035-1038.

54 Prentice, H.M., Moo~e, S.E., Dickson, J.G., Doherty, P. and Walsh, F.S., Nerve growth factor-induced changes in neural cell adhesion molecule (N-CAM) in PC12 cells, EMBO J., 6 (1987) 1859-1863.

87

55 Saadat, S., Stehle, A.D., Lamouroux, A., Mallet, J. and Thoencu, H., Influence of cell-cell contact on levels of tyrosine hydroxylase in cultured bovine adrenal chromaffin cells, J. Biol. Chem., 262 (1987) 13007-13014.

56 Saadat, S. and Thoenen, H., Selective induction of tyrosine hydroxylase by cell-cell contact in bovine adrenal chromaffin cells is mimici~ed by plasma membranes, I. Cell Biol., 103 (1986) ~991-,1997,

57 Saadat, 5. a~d Thoenen, H., Regulation of tyrosine hydroxylase mRNA levels in rat pheochromocytoma PC12 cells by cell-cell contact, "Zxp. Cell Res., 176 (198~) 187-193.

58 Sabban, E.L., Goldstein, M. and Greene, L.A., Regulation of the multiple forms of dopamine beta hy~.,oxylase by nerve growth factor, dexamethasone, and dibuty~l cyclic AMP in the PC12 pheochromocytoma cell line, I. Biol. Chem., 258 (1983) 7819-7823.

59 Schal~ling, M., Stieg, P.E., Lindquist, C., Goldstein, M. and H0kfelt, T., i~apid increase in enzyme and peptide mRNA in sympathetic, ganglia after electrical s~:imulation in humans, Proc. Natl. ,1c~t. Sci. U.S.A., 86 (1989) 4302-4305.

60 Schubert, D. ~nd Witlock, C., Alteration of cellular adhesion by nerve growth factor, Proc. Natl. Acad. Sci. U.S.A., 74 (lr~O7) ~055-4058.

61 Thoenen, H., Kettler, R., Burkard, W. and Saner, A., Neurally mediated control of enzymes involved in the synthesis of norepinephrine; ;~e they regulated as an operational unit?, Arch. Pharmaco~i~, 270 (1971) 1416--!60.

62 Thomas, P.S., Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose, Proc. Natl. Acad. Sci. U.S.A., 77 (I980) 5201-5205.

63 Varon, S.S. and Bunge, R.P~, Trophic mechanisms in the peripheral nervous .~ystem, Annu. Rev. Neurosci., 1 (1978) 327-361.

64 WienshHboum, R.M., Kvetnansky, R., Axelrod, J. and Kopin, l.J., Elevation of s~,rum dopamine beta hydroxylase activity with forced immobilization, Nature New Biol., 230 (197|) 287-288.

65 Wilson, S.P. and Viveros, O.H., Primary cultures of adrenal medullary chromaffin cells in :~emically defined medium, Exp. Cell Res., 133 (1981) 159-169.

66 Wong, V. and Kessler, J.A., Solubilization of membrane factor that stimulates levels of substance P and choline acetyltransfe- rase in sympathetic neurons, Proc. Natl. Acad. $ci. U.S.A., 84 (1987) 8726-8729.